While the amount of textual data available to us is constantly increasing, managing
the texts by human effort is clearly inadequate for the volume and complexity of the
information involved. Consequently, requirement for automated extraction of useful
knowledge from huge amounts of textual data to assist human analysis is apparent.
Text mining (TM) is mostly an automated technique that aims to discover knowledge
from textual data. In this thesis, the notion of text mining, its techniques, applications
are presented. In particular, the study provides the definition and overview of
concepts in text categorization. This would include document representation models,
weighting schemes, feature selection methods, feature extraction, performance
measure and machine learning techniques. The thesis details the functionality of text
mining as a quality improvement tool. It carries out an extensive survey of text
mining applications within service sector and manufacturing industry. It presents two
broad experimental studies tackling the potential use of text mining for the hotel
industry (the comment card analysis), and in automobile manufacturer (miles per
gallon analysis).
Keywords: Text Mining, Text Categorization, Quality Improvement, Service Sector,
Manufacturing Industry.
| Identifer | oai:union.ndltd.org:METU/oai:etd.lib.metu.edu.tr:http://etd.lib.metu.edu.tr/upload/2/12609006/index.pdf |
| Date | 01 November 2007 |
| Creators | J. Mohammad, Mohammad Alkin Cihad |
| Contributors | Weber, Gerhard Wilhelm |
| Publisher | METU |
| Source Sets | Middle East Technical Univ. |
| Language | English |
| Detected Language | English |
| Type | M.S. Thesis |
| Format | text/pdf |
| Rights | To liberate the content for METU campus |
Page generated in 0.0025 seconds